Molecular characterization of the HERV-W env gene in humans and primates: expression, FISH, phylogeny, and evolution.
ABSTRACT We characterized the human endogenous retrovirus (HERV-W) family in humans and primates. In silico expression data indicated that 22 complete HERV-W families from human chromosomes 1-3, 5-8, 10-12, 15, 19, and X are randomly expressed in various tissues. Quantitative real-time RT-PCR analysis of the HERV-W env gene derived from human chromosome 7q21.2 indicated predominant expression in the human placenta. Several copies of repeat sequences (SINE, LINE, LTR, simple repeat) were detected within the complete or processed pseudo HERV-W of the human, chimpanzee, and rhesus monkey. Compared to other regions (5'LTR, Gag, Gag-Pol, Env, 3'LTR), the repeat family has been mainly integrated into the region spanning the 5'LTRs of Gag (1398 bp) and Pol (3242 bp). FISH detected the HERV-W probe (fosWE1) derived from a gorilla fosmid library in the metaphase chromosomes of all primates (five hominoids, three Old World monkeys, two New World monkeys, and one prosimian), but not in Tupaia. This finding was supported by molecular clock and phylogeny data using the divergence values of the complete HERV-W LTR elements. The data suggested that the HERV-W family was integrated into the primate genome approximately 63 million years (Myr) ago, and evolved independently during the course of primate radiation.
- SourceAvailable from: sciencedirect.com[show abstract] [hide abstract]
ABSTRACT: The evolutionary origin and age of the HERV-H family of human endogenous retrovirus-like sequences was investigated in this study. HERV-H elements exist in approximately 900 partially deleted copies and 50-100 more intact forms in humans and Old World monkeys. However, their possible presence in more divergent species is unknown. We have isolated a 1.6-kb genomic DNA segment from the New World monkey marmoset that had been PCR amplified using human HERV-H primers. DNA and protein comparisons and database searches indicate that this marmoset clone is more closely related to human HERV-H elements than to any other sequence, indicating that HERV-H-related sequences do exist in New World monkeys. In contrast to the high copy numbers of deleted elements in Old World primates. Southern blot analysis shows that such elements are present in less than 50 copies in two different species of New World monkey. To estimate evolutionary ages of the common deleted form of the element, a selected DNA segment from the pol region was compared from multiple human HERV-H elements. This comparison suggests that many HERV-H elements of the abundant deleted subfamily integrated approximately 30-35 million years ago. Very similar percentage divergence values between 5' and 3' long terminal repeats of individual elements of the deleted subfamily also suggest that these elements are close in age. These results indicate that HERV-H elements first appeared in the germline prior to the New World/Old World divergence over 40 million years ago. Interestingly, they remained in low numbers in the New World branch while a subfamily underwent a major amplification in Old World primates before the time of divergence of hominoids from Old World monkeys.Virology 12/1995; 213(2):395-404. · 3.37 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Human endogenous retroviral sequences family P (HERV-P) proviral sequences have been located within the human genome. Here, we identify and analyze novel putative structural genes of HERV-P in primates, human tissues, and cancer cells with an aim toward better understanding their evolutionary relationships and transcriptional potential. The expression pattern of HERV-P structural genes indicates that they are actively amplified in human tissues and widely expressed in cancer cells, suggesting a potential role in carcinogenesis. Phylogenetic analyses suggest that the HERV-P family may be divided into two distinct categories that arose during primate evolution via active gene duplication. Taken together, our data provide a better understanding of the dynamic evolutionary features and potential functional roles of the HERV-P gene family.Genomics 02/2007; 89(1):1-9. · 3.01 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: A new human endogenous retroviral family (HERV-W) has been described that is related to multiple sclerosis-associated retrovirus sequences that have been identified in particles recovered from monocyte cultures from patients with multiple sclerosis. Using the polymerase chain reaction approach with a human monochromosomal somatic cell hybrid DNA panel, 15 env fragments of the HERV-W family from chromosomes 1, 3, 4, 5, 6, 7, 12, 14, 17, 20, and X were identified and analyzed. These env fragments showed a high degree of nucleotide sequence similarity (91.6-99.6%) to that of HERV-W. Translation of the env fragments showed no frameshift and termination codons by deletion/insertion or point mutation in some clones (W-1-1, W-3-8, W-4-1, W-7-1, W-14-1, W-17-5, W-20-9, and W-X-3). Phylogenetic analysis of the HERV-W family indicates that the HERV-W env fragments divided into five groups through evolutionary divergence in the primate genome. In group IV, a clone (W-12-2) on chromosome 12 shared 100% sequence identity with a clone (W-17-5) on chromosome 17, suggesting either a retrotransposition or a chromosomal translocation in the last 2 to 5 million years.AIDS Research and Human Retroviruses 06/2001; 17(7):643-8. · 2.71 Impact Factor
Mol. Cells OS, 53-60, July 31, 2008
Molecular Characterization of the HERV-W Env
Gene in Humans and Primates: Expression, FISH,
Phylogeny, and Evolution
Heui-Soo Kim1,*, Dae-Soo Kim2, Jae-Won Huh1, Kung Ahn1, Joo-Mi Yi1, Ja-Rang Lee1, and Hirohisa Hirai3
We characterized the human endogenous retrovirus (HERV-
W) family in humans and primates. få=ëáäáÅç expression data
indicated that 22 complete HERV-W families from human
chromosomes 1–3, 5–8, 10–12, 15, 19, and X are randomly
expressed in various tissues. Quantitative real-time RT-PCR
analysis= of the HERV-W Éåî gene derived from human
chromosome 7q21.2 indicated predominant expression in
the human placenta. Several copies of repeat sequences
(SINE, LINE, LTR, simple repeat) were detected within the
complete or processed pseudo HERV-W of the human,
chimpanzee, and rhesus monkey. Compared to other re-
gions (5′LTR, d~ÖI=d~ÖJmçäI=båîI 3′LTR), the repeat family
has been mainly integrated into the region spanning the
5′LTRs of d~Ö (1398 bp) and mçä (3242 bp). FISH detected
the HERV-W probe (fosWE1) derived from a gorilla fosmid
library in the metaphase chromosomes of all primates (five
hominoids, three Old World monkeys, two New World mon-
keys, and one prosimian), but not in Tupaia. This finding
was supported by molecular clock and phylogeny data us-
ing the divergence values of the complete HERV-W LTR
elements. The data suggested that the HERV-W family was
integrated into the primate genome approximately 63 million
years (Myr) ago, and evolved independently during the
course of primate radiation.
Human endogenous retrovirus (HERV) families are dispersed
over 8% of the human genome. They are believed to have
originated from germ-cell infection with an exogenous retrovirus
during primate evolution (Lower et al., 1996) and their copy
number has increased by retroposition or recombination. Full-
length retroviral sequences can interact with cellular oncogenes
(Varmus, 1982), and retroviral long terminal repeat (LTR) ele-
ments can act as promoters and enhancers of cellular genes
(Lee et al., 2003). Characterization of the HERV elements
should provide information about fundamental cellular activities
and the pathogenesis of multifactorial diseases such as cancer
and autoimmune diseases (Mayer, 2001; Nakagawa et al.,
1997; Yi et al., 2001).
A new HERV-W family has been identified using successive
overlapping cDNA clones from multiple sclerosis patients and
human placenta tissues (Blond et al., 1999; Komurian-Pradel et
al., 1999). qÜÉ=Ö~Ö and éçä open-reading frames (ORFs) are
interrupted by frameshifts and stop codons, and a complete
ORF coding for a putative envelope protein has been identified
(Blond et al.I 1999). The éçä sequence is related to those of
type C oncoviruses, whereas the Éåî sequence is more related
to those of type D oncoviruses (Komurian-Pradel et al.I 1999).
HERV-W structural genes (Ö~ÖI= éçäI= Éåî) have been de-
tected in hominoids and Old World monkeys. It has been esti-
mated that they entered the catarrhine genome more than 25
million years ago (Voisset et al., 2000). However, the HERV-W
LTR-like elements have been detected in hominoids, Old World
monkeys, and New World monkeys (Kim et al., 1999). éçä and
Éåî gene sequences of the HERV-W family in human mono-
chromosomes were identified by cloning and sequencing (Kim,
2001; Kim and Lee, 2001), while the chromosomal distributions
of HERV-W Ö~ÖJI=éêçJI and Éåî-related sequences were found
by Southern blot analysis (Voisset et al., 2000). Here, we as-
sessed the level of expression of the HERV-W Éåî gene in
normal human tissues and determined its chromosomal loca-
tion in primates using fluorescence áå=ëáíì hybridization=(FISH).
MATERIALS AND METHODS
Human RNA samples
Total RNAs from normal human tissues (adrenal gland, bone
marrow, cerebellum, whole brain, fetal brain, fetal liver, heart,
kidney, liver, lung, placenta, prostate, salivary gland, skeletal
muscle, spinal cord, testis, thymus, thyroid grand, trachea, and
uterus) were purchased from Clontech (Supplementary Table
1). Messenger RNAs were extracted using the PolyATtract
mRNA isolation system (Promega).
Blood samples were collected from human (eçãç=ë~éáÉåë),
1Division of Biological Sciences, College of Natural Sciences, Pusan National University, Busan 609-735, Korea. 2PBBRC, Interdisciplinary Research
Program of Bioinformatics, Pusan National University, Busan 609-735, Korea. 3Department of Cellular and Molecular Biology, Primate Research Insti-
tute, Kyoto University, Inuyama 484-8506, Japan.
Received November 11, 2007; revised January 2, 2008; accepted January 7, 2008; published online June 4, 2008
Keywords: FISH, gene expression, HERV-W family, human tissues, primates, RT-PCR
54 The HERV-W Éåî Gene in Humans and Primates
chimpanzee (m~å=íêçÖäçÇóíÉë), gorilla (dçêáää~=Öçêáää~), orangu-
tan (mçåÖç= éóÖã~Éìë), gibbon (eóäçÄ~íÉë= ~Öáäáë), baboon
(m~éáç=Ü~ã~Çêó~ë), Japanese monkey (j~Å~Å~=ÑìëÅ~í~), hy-
brid monkey (African green monkey × patas monkey) (`ÜäçêçJ
ÅÉÄìë=~ÉíÜáçéë × bêóíÜêçÅÉÄìë=é~í~ë), spider monkey (^íÉäÉë=
ÄÉäòÉÄìíÜ), night monkey (^çíìë=íêáîáêÖ~íìë), and lemur (iÉãìê=
Å~íí~). The common tree shrew (qìé~á~=Öäáë) was used as a
Quantitative real-time RT-PCR analysis
Real-time RT-PCR was performed on a Rotor Gene 3000
(Corbett Research) with QuantiTect®SYBR®Green PCR Mas-
ter Mix (Qiagen). Samples were standardized to the GAPDH
mRNA level. Transcripts of the HERV-W Éåî gene were de-
tected by quantitative real-time RT-PCR using primers 5′-
TCATATCTAAGCCCCGCAAC-3′ and 5′-CGTTCCATGTCC-
CCATTTAG-3′ from human HERV-W sequences (GenBank
accession no. NM_014590.3 from chromosome 7q21.2).
Controls showed that the mRNA samples did not contain
genomic DNA. To construct a standard curve samples were
cloned into a standard vector (pGEM®-T Easy Vector). Con-
centrations of the cloning products were measured with a
spectrophotometer (NanoDrop ND-1000), and the corre-
sponding copy numbers were calculated from the formula:
copy number = [6.02 × 1023 (copy/mol) × DNA amount (ng/ul)]/
[plasmid DNA length (bp) × 660 (ng/mol/bp)] (Whelan et al.,
2003). HERV-W Éåî gene transcript copy numbers were
normalized to GAPDH transcript copy numbers. The GAPDH
gene was amplified using the GPH-QS (5′-GAAGATGG-
TGATGGGATTTC-3′) and GPH-QAS (5′-GAAGGTGAAGG-
TCGGAGT-3′) primers from human GAPDH (GenBank ac-
cession no. NM_002046). cDNAs of the HERV-W Éåî and
GAPDH genes were subjected to 30 cycles of amplification
under the following conditions: 94°C denaturing for 10 s, 58°C
annealing for 15 s, and 72°C extension for 15 s.
Data sources and bioinformatic tools
Genome sequences of human, chimpanzee, and rhesus mon-
key were retrieved from NCBI database Build 36.1. HERV-W
elements were identified from the Repbase database and from
the genomic sequences in RepeatMasker (http://ftp.genome.
washington.edu.cgi-bin/RepeatMasker), which uses a cross_
match program to perform perfect sequence alignments (Jurka,
2000). The human transcript data used for studying the expres-
sion of HERV-W in the whole transcriptome were extracted
from the dbEST sequence in the NCBI database.
Analysis of the HERV-W genomic structure
We also performed a bioinformatic analysis of the HERV-W
elements using the RepeatMasker program (http://repeat-
masker.genome.washington.edu) with various repeat consen-
sus sequences from the Repbase Update (Jurka, 2000). The
LTR17 and HERV17 consensus sequences were joined as
consensus sequences. Finally, LTR17-HERV17-LTR17 se-
quences acted as consensus sequences for the HERV-W fam-
ily. The internal retroviral sequences of the HERV-W elements
were constructed by comparing conserved residues as poten-
tial coding regions (Ö~ÖI=éçäI=Éåî) using the BlastX program.
These HERV-W family sequences were divided into four types
(complete HERV-W, processed pseudo HERV-W, truncated
HERV-W, and solitary HERV-W LTR) that were detected by
calculating the RepeatMasker output while reconstructing the
HERV structure of the non-fragmental state before mutations
accumulated. Processed pseudo HERV-W sequences were
defined as incomplete LTRs that start close to the beginning of
the R region of the 5′LTR and end in the 3′ part of the 3′LTR R
Molecular phylogeny of HERV-W LTR elements
Multiple alignment analysis was performed using two different
regions of the 5′LTR and 3′LTR sequences from the complete
HERV-W family (Thompson et al., 1994). The neighbor-joining
phylogenetic tree was obtained with the MEGA3 program
(Kumar et al., 2004). Divergence values were estimated by the
Kimura two-parameter method in the MEGA3 program.
Cell culture and preparation of primate metaphase
Whole blood from various primates was cultured in RPMI1640
medium (Nipro) containing 20% FCS (Gibco BRL), a three-
mitogen mixture (10 μg/ml phytohemagglutinin (Murex Biotech),
3 μg/ml concanavalin A (Sigma), 3 μg/ml lipopolysaccharide
(Sigma), 50 μg/ml streptomycin, and 50 U penicillin. The cells
were grown in a 37°C CO2 incubator for 70 h. After treating the
cells with 50 μg/ml colcemid (Gibco BRL) for 30 min, lympho-
cytes were harvested by centrifugation and fixed with ethanol
and acetic acid (3:1). Details of slide preparation have been
given previously (Hirai et al., 1998; 1999).
cäìçêÉëÅÉåÅÉ= in situ= ÜóÄêáÇáò~íáçå= EcfpeF= çÑ= ebosJt= áå=
The fosWE1 fosmid clone derived from gorilla genomic DNA
(20 μg/μl) was denatured by boiling for 5 min, and labeled using
the BioPrime DNA labeling system with biotin-14-dCTP as
hapten (Invitrogen). For DNA precipitation after labeling, 50 μl
of the labeled probe was added to 690 μl of a mixture of 465 μl
99.5% cold (–20°C) ethanol, 0.5 M NaCl, and 19 μg salmon
testes DNA. The mixture was centrifuged at 21,500 g after
holding at 4°C for 60 min. The labeled probe DNA pellet was
saturated with 50 μl formamide, and a hybridization solution
was made with 10 μl of formamide saturated probe and 15 μl of
hybridization buffer (3:1 30% dextran sulfate: 20× SSC), and
this was then denatured at 72°C for 10 min prior to the initiation
of hybridization. The chromosomal DNA was denatured with
0.05 M NaOH (pH 12.5) in 2× SSC for 4.5 min, followed by
dehydration with 70% and 99.5% ethanol for 5 min each. After
drying, the denatured probe DNA was applied to slides, which
were covered with parafilm and incubated in a moist chamber
at 37°C for 12–16 h. Signals were detected with FITC-avidin
DCS (Vector) after adequate washing, and were observed and
recorded under a fluorescence microscope (Axioplan 2, Zeiss)-
mounted CCD camera connected to a personal computer (Ap-
ple G4) running IPLab imaging software (Scanalytics). These
procedures were modified from a previously described tech-
nique used for BAC DNA clones (Hirai and Hirai, 2004).
få=ëáäáÅç expression analysis and quantitative real-time
RT-PCR analysis of the HERV-W env family
få=ëáäáÅç expression indicated that the 22 complete HERV-W
families in human chromosomes 1–3, 5–8, 10–12, 15, 19, and X
are expressed in various tissues (brain, mammary gland, cere-
brum, skin, testis, eye, embryonic tissue, pancreatic islet, pineal
gland, endocrine, retina, adipose tissue, placenta, and muscle).
The HERV-W element derived from chromosome 12p11.21 is
expressed in pancreatic islets, with 42 hits from the Éåî gene.
The HERV-W element derived from chromosome 7q21.2 is
expressed in placenta, testis, and embryonic tissue, with 224
Heui-Soo Kim et al. 55
Table 1. In silico expression analysis of HERV-W in human normal tissues
d~Ö= mçä= båî
chr1_HS.3 1p22.2 - - - - 1 - --------1- - - 1 -
chr2_HS.48 2q13 - - 1 - - 2 --------3 - - - 3 -
chr2_HS.49 2q32.3 - - - - - - -2------2 - - - 2 -
chr2_HS.51 2p16.2 1 - - - - - --------1- 1 - -
chr3_HS.58 3q11.2 - - - - - - ------3-3 - - - 2 1
chr3_HS.60 3q25.1 - - - - 3 - --------3- - - 2 1
chr5_HS.80 5q12.1 - - - - - - ------1-1- - - 1 -
chr5_HS.84 5q14.3 - - - 1 - - --------1 - - - - 1
chr6_HS.86 6q21 - 1 - - - - ------2-3 - 1 - 2 -
chr6_HS.89 6q21 - - 1 - - - --------1- - 1 - -
chr6_HS.90 6p12.1 - - - - - - ---1----1- - - 1 -
chr6_HS.95 6q24.2 - - - - - - -------11- - - 1 -
chr7_HS.96 7q21.2 - - - - 12 - 5-----207-224 - 11 41 1648
chr8_HS.102 8q12.1 - - - - - - ----1---1- - - 2 -
chr10_HS.8 10q23.1 - - - - - - ------1-1 - - 1 - -
chr11_HS.16 11q14.1 - 2 - - - - ------2-4- - 3 - 1
chr12_HS.21 12p13.31 - - - - - - ------2-2- - - 2 -
chr12_HS.26 12p11.21 - - - - - - -42------42- - - 42 -
chr15_HS.31 15q21.3 - - - - - - --11-2--41 1 1 1 -
chr15_HS.33 15q26.1 - - - - - - ----2---2- - 2 - -
chr19_HS.36 19p12 - - - - - - -------22- 1 - 1 -
chrX_HS.109 Xp11.1 - - 2 - - - --------2 - - 1 1 -
Total 1 3 4 1 16 2 5 44123221833051 15 50 22812
Fig. 1. Quantitative real-time RT-PCR analysis=of the HERV-W Éåî
gene using normal human tissues. The transcript copy number of
the HERV-W Éåî gene was normalized to the GAPDH transcript
copy number for each sample. Each experiment was performed in
hits (11 hits from the Ö~Ö gene, 41 hits from the éçä gene, 164
hits from the Éåî gene, and eight hits from the 3′LTR) (Table 1).
Quantitative real-time RT-PCR analysis=of the HERV-W Éåî
gene derived from human chromosome 7q21.2 indicated pre-
dominant expression in human placenta (Fig. 1). Evidently the
HERV-W Éåî gene is active in various tissues, whereas the
Ö~Ö and éçä genes are silent as a result of mutational events.
Genomic structure and insertion events involving the
Using the RepeatMasker program with consensus sequences
from the Repbase Update, we determined the putative struc-
tures of the consensus sequences, 5′LTR17-HERV17-3′LTR17,
for the HERV-W family (Fig. 2A). The structural genes (Ö~Ö,=éçä,=
Éåî) of the HERV-W family were constructed by comparing
conserved residues using the BlastX program, and confirmed
by domain information from the Pfam HMM database. The
HERV-W family is divided into four types (complete HERV-W,
processed pseudo HERV-W, truncated HERV-W, and solitary
HERV-W LTR), and the different types were detected by calcu-
lating the RepeatMasker output (Fig. 2B). We identified the
complete HERV-W and processed pseudo HERV-W types by
the presence of the intact U3 region of the LTRs on both sides.
The truncated HERV-W type was defined by C-terminal or N-
terminal truncation of the full-length HERV-W type. The solitary
HERV-W LTR type has been shown to posses only an LTR
56 The HERV-W Éåî Gene in Humans and Primates
Fig. 2. Genomic structure of the HERV-W consensus sequences (A) and its four different subtypes (B). The structure was constructed using
the RepeatMasker program with consensus sequences from the Repbase Update. The HERV-W elements are divided into four types (com-
plete HERV-W, processed pseudo HERV-W, truncated HERV-W, and solitary HERV-W LTR), as detected by calculating the RepeatMasker
element without any internal sequences. The majority of retrovi-
ral sequences in the human genome are solitary LTRs that
arose by homologous recombination between the 5′ and 3′
LTRs, resulting in excision of the central region and one LTR.
Using the genome sequences of human, chimpanzee, and
rhesus monkey derived from NCBI database Build 36.1, we
identified the HERV-W elements from the Repbase database
using the RepeatMasker program. Several copies of the repeat
sequences (SINE, LINE, LTR, simple repeat) were detected
within the complete or processed pseudo HERV-Ws of the hu-
man, chimpanzee, and rhesus monkey. SINE elements were
found in abundance (48%). The repeat family is mainly integrated
in the region between the 5′LTR and d~Ö (1398 bp) or mçä (3242
bp) rather than in other regions (5′LTR, d~ÖI= d~ÖJmçäI= båîI
3′LTR) (Table 2; Fig. 2). We also detected complete HERV-W or
processed pseudo HERV-W sequences among the repeat family
of the human, chimpanzee, and rhesus monkey. Many complete
HERV-W sequences were identified within the LINE and LTR
families, while the processed pseudo HERV-W sequences were
abundant among the LINE elements (Table 3).
Chromosomal distribution and evolution of the HERV-W
family in primates
Using the FISH technique, we examined the chromosomal
distribution of the HERV-W family in various primates (five
hominoids, three Old World monkeys, two New World monkeys,
and one prosimian). Using a HERV-W probe (fosWE1) derived
from a gorilla fosmid library, HERV-W elements were detected
in the metaphase chromosomes of all primates except Tupaia
(Fig. 3). We also examined phylogenetic features using the
divergence values of the 5′LTR and 3′LTR sequences from
complete HERV-W LTR elements; the results of this examina-
tion indicated that the HERV-W family became integrated into
the prosimian genome and expanded in the era of Old and
New World monkeys (Fig. 4). Since that event, the 5′LTR and
3′LTR elements have followed independent evolutionary paths
during primate radiation, with a rate of evolution of 0.17% per
Ubiquitous expression of the HERV-W Éåî=gene
Previously the HERV-W Éåî gene was found to be strongly
expressed in the human placenta with two transcripts of 4 and
8 kb, as determined by Northern blot analysis; weaker expres-
sion was also seen in the testis, with the 8-kb transcript pre-
dominating, and no expression was detected in any other hu-
man tissues (Mi et al., 2000). However, in the present work,
quantitative real-time RT-PCR analysis=indicated that the Éåî
gene that was derived from human chromosome 7q21.2 was
ubiquitously expressed in 20 different human tissues, with ro-
bust expression in the placenta (Fig. 1). This expression pattern
was also supported by áå=ëáäáÅç analysis, which indicated that 22
complete HERV-W families from several human chromosomes
are randomly expressed in various human tissues. More spe-
cifically, the HERV-W element (chromosome 7q21.2) is ex-
pressed in the placenta, testis, and embryonic tissue (Table 1).
The Éåî gene also encodes a protein of 80 kDa when glycosy-
lated by áå=îáíêç translation in the presence of canine micro-
somes (Voisset et al., 2000). In our previous sequencing data,
50 clones from normal human tissues and 44 clones from can-
cer cells contained putative amino acid sequences of the
HERV-W Éåî gene, whereas only two clones from cancer cells
contained putative amino acid sequences in the case of the
HERV-W éçä= gene (Yi et al., 2004a), suggesting that the
HERV-W Éåî gene is more active in various tissues and cancer
Relationship between the HERV-W and repeat family
To date, BLAST search for the HERV-W family has identified
Heui-Soo Kim et al. 57
Table 2. Comparative analysis of insertion event of repeat family within HERV-W among human, chimpanzee, and rhesus monkey
SINE LINE LTR Simple repeat
C P C P C P C P
5LTR - - - - - - - - -
LTR-GAG 3 3 1 2 4 3 2 2 20
- - - - - - 2
GAG-POL 1 2 - - - - 1 - 4
POL 5 7 - 1 1 - 3 2 19
ENV 2 - -
3LTR 1 - - - - - - - 1
5LTR - - - - - - - - -
LTR-GAG 4 3 4 2 10 1 3 3 30
GAG 1 - - - - - - - 1
2 - - - - - - 2
POL 3 6 1 - 1 1 3 2 17
ENV 2 - - - - - - - 2
3LTR - - - 1 - - - - 1
5LTR - 1 - - - - - - 1
LTR-GAG 3 4 - 2 5 1 4 4 23
GAG 3 4 - 1 1 - 1 1 11
GAG-POL 1 1 - - - 1 - - 3
POL 4 5 - - - 1 1 1 12
ENV - - - - - - - - -
3LTR - - - - - - - - -
Total 35 38 6 9 22 8 18 15 151
C, Complete HERV-W; P, Processed Psedo HERV-W
Table 3. Insertion of HERV-W within repeat family
SINE LINE LTR DNA Simple repeat
Human 4 11 13 2 - 30
Chimpanzee 6 11 11 2 - 30
Rhesus 3 5 6 - - 14
Human 5 16 5 - 11 37
Chimpanzee 5 12 2 - 5 24
Rhesus 3 5 6 - - 14
140 sequences, representing 39 HERV-W proviruses, 40 full-
length HERV-W retroposons, and 61 truncated HERV-W ret-
rosequences, in the DDBJ/EMBL/GenBank databases (Costas,
2002). The number of identified HERV-W-related fragments, at
least 70 copies for Ö~Ö and 30 copies for Éåî per haploid ge-
nome, is correlated with the previously described increase in
complexity from the Éåî region to the Ö~Ö and éêç regions
(Voisset et al., 2000). As shown in Table 2, the HERV-W ele-
ments were detected in the Repbase database with the Re-
peatMasker program using genome sequences of human,
chimpanzee, and rhesus monkey derived from NCBI database
Build 36.1. Furthermore, the structure of the HERV-W element
was determined using a consensus sequence (Fig. 2). Several
copies of the repeat sequences (SINE, LINE, LTR, simple re-
peat) were identified within the complete or processed pseudo
HERV-W of the human, chimpanzee, and rhesus monkey.
Using these repeat sequences, it was found that multiple inter-
ruptions of the ORF region had occurred during primate evolu-
tion, and such interruptions can affect the genomic stability and
biological functions of structural genes. Within the HERV-W
family, the repeat family has mainly been integrated into the
LTR-d~Ö (1398 bp) and mçä (3242 bp) region rather than other
regions (5′LTR, d~ÖI=d~ÖJmçäI=båî, 3′LTR) (Table 2, Fig. 2).
This indicates that truncation involving insertion mutations of
58 The HERV-W Éåî Gene in Humans and Primates
A B C D
E F G H
I J K L
the abundant SINE mainly occurred in the LTR-d~Ö (1398 bp)
or mçä (3242 bp) region, whereas the Éåî region of the HERV-
W family was well-conserved and may have an important func-
tion. These data are consistent with the biological role of the
Éåî=gene in trophoblast fusion and differentiation (Frendo et al.,
2003; Mi et al., 2000).
We also analyzed the genomic integration sites of HERV-W
using human, chimpanzee, and rhesus monkey genomes.
Interestingly, the processed pseudo HERV-W sequences were
more abundant in the LINE than in the SINE or LTR elements
(Table 3). The distribution of complete HERV-W and pseudo
HERV-W copies was biased towards the GC-poor regions of
the genome, and pseudo HERV-W had a strong tendency to be
found in GC-poor regions. This bias is typical for LINEs, young
^äìë, and processed pseudogenes, and is probably linked to
the AT-rich insertion motif of these elements (Pavlíek et al.,
2001). Therefore, it is conceivable that coexpression of LINEs
and HERV-W elements facilitates the formation of processed
pseudogenes in the HERV-W family. Processed pseudo
HERV-W may arise by multiple LINE-mediated retrotransposi-
tion of retroviral mRNAs. Involvement of the LINE1 machinery
in the genomic integration of retroviral mRNAs is suggested by
the existence of a considerable proportion of genomic RNA
sequestered by the LINE1 elements (Costas, 2002). These
findings suggest that parasitism in the LINE1 element facilitated
the proliferation of HERV-W retrosequences.
Chromosomal distribution and evolutionary features of the
Most HERV families entered the genome between 10 and 50
million years ago during the course of primate evolution (Jurka,
2000). Their proliferation may have been a result of germ-line
reinfection rather than retrotransposition in the Åáë or íê~åë posi-
tion (Belshaw et al., 2004). Since HERV-W structural genes
have been detected in hominoids and Old World monkeys, it
has been estimated that they entered catarrhine genomes
more than 25 million years ago (Voisset et al., 2000). However,
the HERV-W LTR-like elements have been detected in homi-
noids, Old World monkeys, and New World monkeys (Kim et
al., 1999). In the present study, we determined the chromoso-
mal distribution of the HERV-W family in various primates using
a HERV-W probe (fosWE1) derived from the gorilla fosmid
library. The fosWE1 probe clearly detected sequences in the
metaphase chromosomes of all primates except Tupaia (Fig. 3).
A similar phenomenon was observed in the HERV-P éçä gene,
as detected in apes, Old and New World monkeys and
prosimians, by PCR amplification, whereas the Éåî gene was
not detected in prosimians (Yi et al., 2007). In the case of the
HERV-S éçä gene, the PCR products were generated by sam-
ples of hominoids, Old World monkeys, and New World mon-
keys, but not of prosimians, which indicates that the HERV-S
family was integrated into the primate genome prior to the di-
vergence of Old and New World monkeys (Yi et al., 2004b).
We also examined phylogenetic features using the diver-
gence values of the 5′LTR and 3′LTR sequences from com-
plete HERV-W LTR elements because sequence divergences
between the LTRs provide a unique and important source of
phylogenetic information. According to the mechanism of re-
verse transcription, the two LTRs must be identical. Over time,
they tend to diverge in sequence as a result of substitutions,
insertions, and deletions acquired during cellular DNA replica-
tion, so that the divergence between the two LTRs can serve
as a molecular clock to reveal the evolution of the HERV family
(Mager and Freeman, 1995). As shown in Fig. 4, the HERV-W
family was integrated into the prosimian genome and expanded
in the era of Old and New World monkeys, as determined by
use of the molecular clock. After that event, both 5′LTR and
3′LTR elements were processed via independent evolutionary
paths during primate radiation, with an evolutionary rate of
0.17% per million years (Myr). Various evolutionary rates have
been detected during primate evolution; these varied between
different members of the HERV family. The estimated rates
Fig. 3. FISH analysis of the
HERV-W family with a fosmid
clone (fosWE1) derived from
gorilla genomic DNA on meta-
phase chromosomes from (A)
human, (B) chimpanzee, (C)
gorilla, (D) orangutan, (E)
gibbon, (F) baboon, (G) Japa-
nese monkey, (H) hybrid
monkey (African green mon-
key × patas monkey), (I) spi-
der monkey, (J) night monkey,
and (K) lemur, (L) The com-
mon tree shrew was used as
an outgroup of non-primates.
Yellow bands indicate signals
of the HERV-W family.
Heui-Soo Kim et al. 59
Fig. 4. Phylogenetic tree based on 5′LTR and 3′LTR sequences derived from complete HERV-W LTR elements in the human genome. The
5′LTR and 3′LTR sequences do not cluster together. The tree was constructed using the neighbor-joining method. Branch lengths are propor-
tional to the divergence between the taxa. The putative molecular clock was calculated, and is indicated under the tree using an evolutionary
rate of 0.17% per Myr.
range from 0.2% per Myr for HERV-H LTRs (Anderssen et al.,
1997), to 0.16% per Myr for ERV9 LTRs (Costas and Naveira,
2000) and 0.13% per Myr for HERV-K LTRs (Lebedev et al.,
2000). The HERV-W LTR elements gave a similar value to that
of the ERV-9 LTRs. These elements have a sister relationship
that is consistent with molecular phylogeny data (Tristem,
2000). Compared to the other LTR elements, the LTRs of the
HERV-W and ERV9 cluster together indicating that closely-
related phylogenetic groups evolved at a similar evolutionary
rate during the course of primate radiation.
This work was supported by the Korea Science and Engineer-
ing Foundation (KOSEF) grant funded by the Korea Govern-
ment (Ministry of Science and Technology) (No. R01-2007-000-
20035-0), the 21st Century COE (A14), and the Global COE
(A06) (Japan). H.S.K. received the opportunity to work as a
foreign associate professor for three months at the Primate
Research Institute, Kyoto University. We thank Yuriko Hirai for
her technical assistance with FISH.
Anderssen, S., Sjottem, E., Svineng, G., and Johansen, T. (1997).
Comparative analysis of LTRs of the ERV-H family of primate-
specific retrovirus-like elements isolated from marmoset, African
green monkey, and man. Virology OPQ, 14-30.
Belshaw, R., Pereira, V., Katzourakis, A., Talbot, G., Paces, J., Burt,
A., and Tristem, M. (2004). Long-term reinfection of the human
genome by endogenous retroviruses. Proc. Natl. Acad. Sci.
USA NMN, 4894-4899.
Blond, J.L., Beseme, F., Duret, L., Bouton, O., Bedin, F., Perron, H.,
Mandrand, B., and Mallet, F. (1999). Molecular characterization
and placental expression of HERV-W, a new human endoge-
nous retrovirus family. J. Virol. TP, 1175-1185.
Costas, J. (2002). Characterization of the intragenomic spread of
the human endogenous retrovirus family HERV-W. Mol. Biol.
Evol. NV, 526-533.
Costas, J., and Naveira, H. (2000). Evolutionary history of the hu-
man endogenous retrovirus family ERV9. Mol. Biol. Evol. NT,
Frendo, J.L., Olivier, D., Cheynet, V., Blond, J.L., Bouton, O., Vi-
daud, M., Rabreau, M., Evain-Brion, D., and Mallet, F. (2003).
Direct involvement of HERV-W Éåî glycoprotein in human tro-
phoblast cell fusion and differentiation. Mol. Cell. Biol.=OP, 3566-
Hirai, H., and Hirai, Y. (2004). FISH mapping for helminth genome.
In Methods in Molecular Biology, Parasite Genomics Protocols,
S.E. Melville, ed. (Totowa, USA: Human Press). OTM, 379-394.
60 The HERV-W Éåî Gene in Humans and Primates
Hirai, H., Hasegawa, Y., Kawamoto, Y., and Tokita, E. (1998). Tan-
dem duplication of nucleolus organizer region (NOR) in the
Japanese macaque, j~Å~Å~=ÑìëÅ~í~=ÑìëÅ~í~. Chrom. Res.=S,
Hirai, H., Taguchi, T., and Godwin, A.K. (1999). Genomic differen-
tiation of 18S ribosomal DNA and β-satellite DNA in the homi-
noid and its evolutionary aspects. Chrom. Res. T, 531-540.
Jurka, J. (2000). Repbase update: a database and an electronic
journal of repetitive elements. Trends Genet. NS, 418-420.
Kim, H.-S., Takenaka, O., and Crow, T.J. (1999). Isolation and
phylogeny of endogenous retrovirus sequences belonging to the
HERV-W family in primates. J. Gen. Virol.=UM, 2613-2619.
Kim, H.-S. (2001). Sequence and phylogeny of HERV-W éçä frag-
ments. AIDS Res. Hum. Retroviruses 17, 1665-1671.
Kim, H.-S., and Lee, W.-H. (2001). Human endogenous retrovirus
HERV-W family: chromosomal localization, identification, and
phylogeny. AIDS Res. Hum. Retroviruses NT, 643-648.
Komurian-Pradel, F., Paranhos-Baccala, G., Bedin, F. Ounanian-
Paraz, A., Sodoyer, M., Ott, C., Rajoharison, A., Garcia, E., Mal-
let, F., Mandrand, B., et al. (1999). Molecular cloning and char-
acterization of MSRV-related sequences associated with retrovi-
rus-like particles. Virology OSM, 1-9.
Kumar, S., Tamura, K., and Nei, M. (2004). MEGA3: Integrated
software for molecular evolutionary genetics analysis and se-
quence alignment. Brief Bioinform. R, 150-163.
Lebedev, Y.B., Belonovitch, O.S., Zybrova, N.V., Khil, P.P.,
Kurdyukov, S.G., Vinogradova, T.V., Hunsmann, G., and Sverd-
lov, E.D. (2000). Differences in HERV-K LTR insertions in
orthologous loci of humans and great apes. Gene OQT, 265-277.
Lee, W.J., Kwun, H.J., Kim, H.-S., and Jang, K.L. (2003). Activation
of the human endogenous retrovirus W long terminal repeat by
herpes simplex virus type 1 immediate early protein 1. Mol. Cells
Lower, R., Lower, J., and Kurth, R. (1996). The viruses in all of us:
characteristics and biological significance of human endogenous
retrovirus sequences. Proc. Natl. Acad. Sci. USA=VP, 5177-5184.
Mager, D.L., and Freeman, J.D. (1995). HERV-H endogenous
retroviruses: presence in the New World branch but amplifica-
tion in the Old World primate lineage. Virology NM, 395-404.
Mayer, J. (2001). Status of HERV in human cells: expression and
coding capacity of human proviruses. Dev. BiolK NMS, 439-441.
Mi, S., Lee, X., Li, X., Veldman, G.M., Finnerty, H., Racie, L.,
LaVallie, E., Tang, X.Y., Edouard, P., Howes, S., et al. (2000).
Syncytin is a captive retroviral envelope protein involved in hu-
man placental morphogenesis. Nature QMP, 785-789.
Nakagawa, K., Brusic, V., McColl, G., and Harrison, L.C. (1997).
Direct evidence for the expression of multiple endogenous retro-
viruses in the synovial compartment in rheumatoid arthritis. Ar-
thritis Rheum. QM, 627-638.
Pavlíek, A., Jabbari, K., Paes, J., Paes, V., Hejnar, J., and Bernardi,
G. (2001). Similar integration but different stability of Alus and
LINEs in the human genome. Gene OTS, 39-45.
Perron, H., Garson, J.A., Bedin, F. Beseme, F., Paranhos-Baccala,
G., Komurian-Pradel, F., Mallet, F., Tuke, P.W., Voisset, C.,
Blond, J.L., et al. (1997). Molecular identification of a novel ret-
rovirus repeatedly isolated from patients with multiple sclerosis.
Proc. Natl. Acad. Sci. USA=VQ, 7583-7588.
Thompson, J.D., Higgins, D.G., and Gibson, T.J. (1994). CLUSTAL
W: Improving the sensitivity of progressive multiple sequence
alignment through sequence weighting, position-specific gap
penalties and weight matrix choice. Nucleic Acids Res. OO,
Tristem, M. (2000). Identification and characterization of novel hu-
man endogenous retrovirus families by phylogenetic screening
of the human genome mapping project database. J. Virol. TQ,
Varmus, H.E. (1982). Form and function of retroviral proviruses.
Science ONS, 812-820.
Voisset, C., Bouton, O., Bedin, F., Duret, B., Mandrand, B., Mallet
F., and Paranhos-Baccala, G. (2000). Chromosomal distribution
and coding capacity of the human endogenous retrovirus
HERV-W family. AIDS Res. Hum. Retroviruses NS, 731-740.
Whelan, J.A., Russel, N.B., and Whelan, M.A. (2003). A method for
the absolute quantification of cDNA using real time PCR. J. Im-
munol. Meth. OTU, 261-269.
Yi, J.M., Kim, H.M., and Kim, H.-S. (2001). Molecular cloning and
phylogenetic analysis of the human endogenous retrovirus
HERV-K long terminal repeat elements in various cancer cells.
Mol. Cells NO, 137-141.
Yi, J.M., Kim, H.M., and Kim, H.-S. (2004a). Expression of the hu-
man endogenous retrovirus HERV-W family in various human
tissues and cancer cells. J. Gen. Virol. UR, 1203-1210.
Yi, J.M., Kim, T.H., Huh, J.W. Park, K.S., Jang, S.B., Kim, H.M., and
Kim, H.S. (2004b). Human endogenous retroviral elements be-
longing to the HERV-S family from human tissues, cancer cells,
and primates: expression, structure, phylogeny and evolution.
Gene PQO, 283-292.
Yi, J.M., Schuebel, K., and Kim, H.-S. (2007). Molecular genetic
analyses of human endogenous retroviral elements belonging to
the HERV-P family in primates, human tissues, and cancer cells.
Genomics UV, 1-9.